TensorFlow25: 使用深度学习做阅读理解+完形填空

记的在学生时代，英语考试有这么一种类型的题，叫：阅读理解。首先让你读一段洋文材料，然后回答一些基于这个洋文材料提的问题。

我先给你出一道阅读理解

Big Panda learned to code when he was 21. He live in China and have no life, feel like a big loser. But here is one thing Panda want you to remember…it´s never too late! You can do anything if you put your heart on it!

____ is the loser.（下划线处该填什么呢？）

我出的这道填空题，对人来说轻而易举，但是要让机器回答就很难了。机器阅读和理解人类语言是非常有挑战性的。

本帖就使用TensorFlow练习一个阅读理解，看看准确率能到什么程度。

使用的数据集

• https://research.fb.com/projects/babi/
• http://cs.nyu.edu/~kcho/DMQA/

本帖只使用”非死不可”提供的《Children’s Book Test》数据集。

数据预处理

import re
import random
import ast
import itertools
import pickle
import numpy as np

train_data_file = './CBTest/data/cbtest_NE_train.txt'
valid_data_file = './CBTest/data/cbtest_NE_valid_2000ex.txt'

def preprocess_data(data_file, out_file):
    # stories[x][0]  tories[x][1]  tories[x][2]
    stories = []
    with open(data_file) as f:
        story = []
        for line in f:
            line = line.strip()
            if not line:
                story = []
            else:
                _, line = line.split(' ', 1)
                if line:
                    if '\t' in line:
                        q, a, _, answers = line.split('\t')
                        # tokenize
                        q = [s.strip() for s in re.split('(\W+)+', q) if s.strip()]
                        stories.append((story, q, a))
                    else:
                        line = [s.strip() for s in re.split('(\W+)+', line) if s.strip()]
                        story.append(line)

    samples = []
    for story in stories:
        story_tmp = []
        content = []
        for c in story[0]:
            content += c
        story_tmp.append(content)
        story_tmp.append(story[1])
        story_tmp.append(story[2])

        samples.append(story_tmp)

    random.shuffle(samples)
    print(len(samples))

    with open(out_file, "w") as f:
        for sample in samples:
            f.write(str(sample))
            f.write('\n')

preprocess_data(train_data_file, 'train.data')
preprocess_data(valid_data_file, 'valid.data')

# 创建词汇表
def read_data(data_file):
    stories = []
    with open(data_file) as f:
        for line in f:
            line = ast.literal_eval(line.strip())
            stories.append(line)
    return stories


stories = read_data('train.data') + read_data('valid.data')

content_length = max([len(s) for s, _, _ in stories])
question_length = max([len(q) for _, q, _ in stories])
print(content_length, question_length)

vocab = sorted(set(itertools.chain(*(story + q + [answer] for story, q, answer in stories))))
vocab_size = len(vocab) + 1
print(vocab_size)
word2idx = dict((w, i + 1) for i,w in enumerate(vocab))
pickle.dump((word2idx, content_length, question_length, vocab_size), open('vocab.data', "wb"))

# From keras 补齐
def pad_sequences(sequences, maxlen=None, dtype='int32',
                  padding='post', truncating='post', value=0.):
    lengths = [len(s) for s in sequences]

    nb_samples = len(sequences)
    if maxlen is None:
        maxlen = np.max(lengths)

    # take the sample shape from the first non empty sequence
    # checking for consistency in the main loop below.
    sample_shape = tuple()
    for s in sequences:
        if len(s) > 0:
            sample_shape = np.asarray(s).shape[1:]
            break

    x = (np.ones((nb_samples, maxlen) + sample_shape) * value).astype(dtype)
    for idx, s in enumerate(sequences):
        if len(s) == 0:
            continue  # empty list was found
        if truncating == 'pre':
            trunc = s[-maxlen:]
        elif truncating == 'post':
            trunc = s[:maxlen]
        else:
            raise ValueError('Truncating type "%s" not understood' % truncating)

        # check `trunc` has expected shape
        trunc = np.asarray(trunc, dtype=dtype)
        if trunc.shape[1:] != sample_shape:
            raise ValueError('Shape of sample %s of sequence at position %s is different from expected shape %s' %
                             (trunc.shape[1:], idx, sample_shape))

        if padding == 'post':
            x[idx, :len(trunc)] = trunc
        elif padding == 'pre':
            x[idx, -len(trunc):] = trunc
        else:
            raise ValueError('Padding type "%s" not understood' % padding)
    return x

# 转为向量
def to_vector(data_file, output_file):
    word2idx, content_length, question_length, _ = pickle.load(open('vocab.data', "rb"))

    X = []
    Q = []
    A = []
    with open(data_file) as f_i:
        for line in f_i:
            line = ast.literal_eval(line.strip())
            x = [word2idx[w] for w in line[0]]
            q = [word2idx[w] for w in line[1]]
            a = [word2idx[line[2]]]

            X.append(x)
            Q.append(q)
            A.append(a)

    X = pad_sequences(X, content_length)
    Q = pad_sequences(Q, question_length)

    with open(output_file, "w") as f_o:
        for i in range(len(X)):
            f_o.write(str([X[i].tolist(), Q[i].tolist(), A[i]]))
            f_o.write('\n')

to_vector('train.data', 'train.vec')
to_vector('valid.data', 'valid.vec')


"""
# to_word
word2idx, content_length, question_length, _ = pickle.load(open('vocab.data', "rb"))

def get_value(dic,value):
    for name in dic:
        if dic[name] == value:
            return name

with open('train.vec') as f:
    for line in f:
        line = ast.literal_eval(line.strip())
        for word in line[0]:
            print(get_value(word2idx, word))
"""

生成的文件：vocab.data词汇表、train.vec、valid.vec数据的向量表示。

训练

https://arxiv.org/pdf/1606.02245v4.pdf

import tensorflow as tf
import pickle
import numpy as np
import ast
from collections import defaultdict

train_data = 'train.vec'
valid_data = 'valid.vec'

word2idx, content_length, question_length, vocab_size = pickle.load(open('vocab.data', "rb"))
print(content_length, question_length, vocab_size)

batch_size = 64

train_file = open(train_data)
def get_next_batch():
    X = []
    Q = []
    A = []
    for i in range(batch_size):
        for line in train_file:
            line = ast.literal_eval(line.strip())
            X.append(line[0])
            Q.append(line[1])
            A.append(line[2][0])
            break

    if len(X) == batch_size:
        return X, Q, A
    else:
        train_file.seek(0)
        return get_next_batch()

def get_test_batch():
    with open(valid_data) as f:
        X = []
        Q = []
        A = []
        for line in f:
            line = ast.literal_eval(line.strip())
            X.append(line[0])
            Q.append(line[1])
            A.append(line[2][0])
        return X, Q, A


X = tf.placeholder(tf.int32, [batch_size, content_length])   # 洋文材料
Q = tf.placeholder(tf.int32, [batch_size, question_length])  # 问题
A = tf.placeholder(tf.int32, [batch_size])                   # 答案

# drop out
keep_prob = tf.placeholder(tf.float32)

def glimpse(weights, bias, encodings, inputs):
    weights = tf.nn.dropout(weights, keep_prob)
    inputs = tf.nn.dropout(inputs, keep_prob)
    attention = tf.transpose(tf.matmul(weights, tf.transpose(inputs)) + bias)
    attention = tf.batch_matmul(encodings, tf.expand_dims(attention, -1))
    attention = tf.nn.softmax(tf.squeeze(attention, -1))
    return attention, tf.reduce_sum(tf.expand_dims(attention, -1) * encodings, 1)

def neural_attention(embedding_dim=384, encoding_dim=128):
    embeddings = tf.Variable(tf.random_normal([vocab_size, embedding_dim], stddev=0.22), dtype=tf.float32)
    tf.contrib.layers.apply_regularization(tf.contrib.layers.l2_regularizer(1e-4), [embeddings])

    with tf.variable_scope('encode'):
        with tf.variable_scope('X'):
            X_lens = tf.reduce_sum(tf.sign(tf.abs(X)), 1)
            embedded_X = tf.nn.embedding_lookup(embeddings, X)
            encoded_X = tf.nn.dropout(embedded_X, keep_prob)
            gru_cell = tf.nn.rnn_cell.GRUCell(encoding_dim)
            outputs, output_states = tf.nn.bidirectional_dynamic_rnn(gru_cell, gru_cell, encoded_X, sequence_length=X_lens, dtype=tf.float32, swap_memory=True)
            encoded_X = tf.concat(2, outputs)
        with tf.variable_scope('Q'):
            Q_lens = tf.reduce_sum(tf.sign(tf.abs(Q)), 1)
            embedded_Q = tf.nn.embedding_lookup(embeddings, Q)
            encoded_Q = tf.nn.dropout(embedded_Q, keep_prob)
            gru_cell = tf.nn.rnn_cell.GRUCell(encoding_dim)
            outputs, output_states = tf.nn.bidirectional_dynamic_rnn(gru_cell, gru_cell, encoded_Q, sequence_length=Q_lens, dtype=tf.float32, swap_memory=True)
            encoded_Q = tf.concat(2, outputs)

    W_q = tf.Variable(tf.random_normal([2*encoding_dim, 4*encoding_dim], stddev=0.22), dtype=tf.float32)
    b_q = tf.Variable(tf.random_normal([2*encoding_dim, 1], stddev=0.22), dtype=tf.float32)
    W_d = tf.Variable(tf.random_normal([2*encoding_dim, 6*encoding_dim], stddev=0.22), dtype=tf.float32)
    b_d = tf.Variable(tf.random_normal([2*encoding_dim, 1], stddev=0.22), dtype=tf.float32)
    g_q = tf.Variable(tf.random_normal([10*encoding_dim, 2*encoding_dim], stddev=0.22), dtype=tf.float32)
    g_d = tf.Variable(tf.random_normal([10*encoding_dim, 2*encoding_dim], stddev=0.22), dtype=tf.float32)

    with tf.variable_scope('attend') as scope:
        infer_gru = tf.nn.rnn_cell.GRUCell(4*encoding_dim)
        infer_state = infer_gru.zero_state(batch_size, tf.float32)
        for iter_step in range(8):
            if iter_step > 0:
                scope.reuse_variables()

            _, q_glimpse = glimpse(W_q, b_q, encoded_Q, infer_state)
            d_attention, d_glimpse = glimpse(W_d, b_d, encoded_X, tf.concat_v2([infer_state, q_glimpse], 1))

            gate_concat = tf.concat_v2([infer_state, q_glimpse, d_glimpse, q_glimpse * d_glimpse], 1)

            r_d = tf.sigmoid(tf.matmul(gate_concat, g_d))
            r_d = tf.nn.dropout(r_d, keep_prob)
            r_q = tf.sigmoid(tf.matmul(gate_concat, g_q))
            r_q = tf.nn.dropout(r_q, keep_prob)

            combined_gated_glimpse = tf.concat_v2([r_q * q_glimpse, r_d * d_glimpse], 1)
            _, infer_state = infer_gru(combined_gated_glimpse, infer_state)

    return tf.to_float(tf.sign(tf.abs(X))) * d_attention

def train_neural_attention():
    X_attentions = neural_attention()
    loss = -tf.reduce_mean(tf.log(tf.reduce_sum(tf.to_float(tf.equal(tf.expand_dims(A, -1), X)) * X_attentions, 1) + tf.constant(0.00001)))

    optimizer = tf.train.AdamOptimizer(learning_rate=0.001)
    grads_and_vars = optimizer.compute_gradients(loss)
    capped_grads_and_vars = [(tf.clip_by_norm(g, 5), v) for g,v in grads_and_vars]
    train_op = optimizer.apply_gradients(capped_grads_and_vars)

    saver = tf.train.Saver()
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())

        # writer = tf.summary.FileWriter()
        # 恢复前一次训练
        ckpt = tf.train.get_checkpoint_state('.')
        if ckpt != None:
            print(ckpt.model_checkpoint_path)
            saver.restore(sess, ckpt.model_checkpoint_path)
        else:
            print("没找到模型")

        for step in range(20000):
            train_x, train_q, train_a = get_next_batch()
            loss_, _ = sess.run([loss, train_op], feed_dict={X:train_x, Q:train_q, A:train_a, keep_prob:0.7})
            print(loss_)

            # 保存模型并计算准确率
            if step % 1000 == 0:
                path = saver.save(sess, 'machine_reading.model', global_step=step)
                print(path)

                test_x, test_q, test_a = get_test_batch()
                test_x, test_q, test_a = np.array(test_x[:batch_size]), np.array(test_q[:batch_size]), np.array(test_a[:batch_size])
                attentions = sess.run(X_attentions, feed_dict={X:test_x, Q:test_q, keep_prob:1.})
                correct_count = 0
                for x in range(test_x.shape[0]):
                    probs = defaultdict(int)
                    for idx, word in enumerate(test_x[x,:]):
                        probs[word] += attentions[x, idx]
                    guess = max(probs, key=probs.get)
                    if guess == test_a[x]:
                        correct_count += 1
                print(correct_count / test_x.shape[0])

train_neural_attention()

我只想说，这个东西比我水平高！

• Attention-over-Attention Neural Networks for Reading Comprehension
• Iterative Alternating Neural Attention for Machine Reading